Table A.4 Developments in service delivery: summary of articles examined for tertiary services

ID and authors	Research setting and year of study	Study design	Sample size	Topic of document	Key results and conclusions	Level of evidence	Quality score (AGREE)(%)
Ferguson et al⁷³³	UK, NHS (1997)	Review of evidence and recommendations	NA	Concentration and choice of hospital services	Specialisation has important implications for service configuration—a small trust may need to employ more consultants to provide the required skills. There is no compelling reason to believe that further concentration of hospital services will improve efficiency or clinical outcomes. In consideration of the negative effects of concentrated access and utilisation, the implications for disadvantaged groups (for example, smaller departments with low funding) should not be overlooked	1	83
Sowden⁷³⁶	International (1995; literature taken between 1985 and 1994)	Systematic review	Over 100 studies	Relationship between volume and quality of health care	There is little evidence as to whether merging hospitals to create larger units will result in a change of outcomes. Owing to this uncertainty, caution should be exercised in using research literature to justify policies of reorganisation of healthcare delivery. The main recommendation is that policy makers should be cautious when invoking the assumed improvements in outcome achieved by volume as a key argument for centralisation of services	1	75
Grilli et al⁷³⁴	International (1980 onwards)	Literature review	47 Papers	The impact of specialised services	The impact of specialised cancer care has been poorly assessed. It is not possible through existing studies to define an optimal configuration of health services for oncology	1	73
Bachmann et al⁷³⁰	England and Wales (1996–97)	Cohort follow-up study	782 Patients	Specialised care for pancreatic cancer	Specialisation of pancreatic cancer care in hospitals was associated with longer survival. Patients referred to less specialised doctors and hospitals were less likely to be investigated thoroughly. A concentration of pancreatic cancer care into higher volume hospitals is likely to improve survival even among patients with incurable disease	2+	73
Bachmann et al⁷²³	Southwest England (1996–97)	Analysis of routine data	1512 Patients	Analysis of specialised GI cancer care	Lower mortality is associated with more specialised care; the concentration of cancer care in the UK is supported	2+	72
CRD⁵⁵⁷	UK, NHS (2000)	Commentary; review of evidence	Around 230 articles	Management of upper GI cancers	There is evidence for each type of GI cancer that treatment in hospitals which manage larger numbers of these patients, and/or by clinicians who see larger numbers, leads to better outcomes. There should be clearly documented policies for patient referral between hospitals, and for the processes by which clinicians seek advice from specialist treatment teams about the management of patients for whom referral may not be appropriate	2+	66
CRD⁵⁵⁹	UK (1997)	Commentary; review of evidence	Around 130 articles	Management of colorectal cancers	An American study found that trained nurses were as likely to discover cancers by sigmoidoscopy as gastroenterologists; patients were more willing to return for a repeat procedure after examination by a nurse. There is some evidence that volume of activity and specialisation may be associated with better surgical technique or practice	2+	59
CRD⁵⁶⁰	UK (2004)	Commentary; review of evidence	Around 190 articles	Management of colorectal cancers	Six systematic reviews and a number of more recent primary studies were consistent in showing evidence that for rectal cancer at least, higher patient volumes and greater specialisation among surgeons were associated with much better outcomes, lower surgical complication rates, decreased local recurrence, lower colostomy rates, and improved survival	2+	59
Senapati et al⁷²⁶	UK (2003)	Survey of clinicians	583 Survey responses	Surgical management of cholelithiasis	Management of cholelithiasis in patients with acute biliary pancreatitis in the UK remains suboptimal. Moreover, only a minority of surgeons offer patients presenting with acute cholecystitis the benefits of early laparoscopic cholecystectomy. The management of acute biliary disease might be improved if these cases were concentrated in the hands of surgeons with upper GI/hepato-pancreato-biliary interest and those who perform laparoscopic cholecystectomy regularly	2−	73
Bachmann et al⁷¹⁶	Southwest England and south Wales (1996–97)	Cohort follow-up study	2294 Patients	Costs of GI cancer care and specialisation	A greater concentration of specialised hospital cancer care will cost more; doctors’ specialisation is as important as hospital specialisation for the effectiveness of cancer care	2−	73
Parks et al⁷²⁹	Scotland (1993–97)	Analysis of routine data	2794 Patients	Benefits (or otherwise) of specialised GI cancer care	Surgically treated patients with pancreatic cancer are likely to fare better when managed by specialist pancreatic surgeons, or clinicians with an interest in this field. Specialisation and concentration of cancer care has major implications for service delivery	2−	68
Neoptolemos et al⁷³¹	UK (1995–96)	Survey	1026 Cases	UK survey of specialist pancreatic units	The data argue strongly in favour of concentrating pancreatic surgery into specialised units, and a higher referral rate from gastroenterologists exclusively to such units should be encouraged	2−	64
McKiernan et al³⁷³	UK (1993–95)	Analysis of patient data; survey	93 Cases	Biliary atresia	Biliary atresia is a rare but important condition and this study confirms that outcome is better if surgical management is done in centres with experience. Throughout the study period referral to surgical centres appeared to be on broadly geographical grounds without any other obvious selection criterion. With this information the major challenge in the future management of biliary atresia will be to ensure that surgical management is rationalised to fewer centres. The need to facilitate the concentration of medical and surgical expertise, junior surgical training, and development of support services has received general acceptance in other areas of paediatric specialist practice. The data presented here suggest to us that children with biliary atresia should be managed in centres with a caseload of more than five cases annually to ensure a better outcome	2−	64
Davenport et al⁷²⁷	England and Wales (1999–2002)	Analysis of routine data	148 Patients	Management of biliary atresia	Early results indicate that improvements in treatment and overall survival of biliary atresia can be achieved nationally through a centralisation of care to supraregional centres	2−	59
Hutchins et al⁷³⁵	UK (1992–98)	Retrospective study; analysis of routine data	65 Patients	Pancreatic surgery in a district general hospital	Compared with specialised pancreatic centres, pancreatic surgery can be performed safely in general district hospitals with low mortality and morbidity, and good long term outcomes. This workload should be undertaken with a dedicated surgeon with the necessary ancillary facilities	2−	55
Williams⁷²²	UK (2004)	Survey	34 Liver centres	Provision of specialist liver services	The survey showed substantial deficiencies in staffing levels, particularly of consultant hepatologists and specialist nurses. Of the various bottlenecks identified in support services, lack of expansion in radiological facilities was the greatest. The failure to provide dedicated beds for hepatology services and sufficient numbers of general and specialist outpatient clinics, as well as the waiting times in many of the centres, contribute to major limitations in service provision. Increasing the number of transplant centres would be one way of enhancing the level of provision of liver services generally	2−	52
Duxbury et al⁷²⁵	UK (1999–2000)	Analysis of routine data	211 Patients	Management of colorectal cancer	This study found striking variation in practice within one hospital in Devon. Colorectal specialists were more likely to conform to best practice guidelines, involve the colorectal nurse specialists to a greater extent with patients with rectal cancer, and perform more extensive lymphadenectomy in rectal cancer surgery. Formal involvement of the colorectal specialist nurse further improves the opportunity for patient information	2−	48
Majeed and Price⁷²⁸	UK (1997–2002)	Analysis of service delivery	615 Patients	Resource and manpower calculations for hepatobiliary surgical services	A centralisation of hepatobiliary surgical services is supported. We estimate that a minimum of two full-time specialist hepatobiliary surgeons with appropriate ancillary support is required for a typical population of two million people in the UK	3	64
Andren-Sandberg and Neoptolemos⁷³²	International, with emphasis on UK (2002)	Literature review	96 Articles	Pancreatic cancer	In the UK, pancreatic surgery must be concentrated into regional centres ideally serving catchment areas of 2–4 million. Smaller district hospitals will have the role of determining provisional diagnosis and stages. It seems highly likely that regionalisation of pancreatic cancer surgery will be adopted	3	45
Senate of Surgery of GB and Ireland⁷²¹	UK, NHS (2004)	Expert commentary and recommendations	NA	Configuration of healthcare services	Where appropriate, clinical care should be provided locally, but patients should be moved to a centre of excellence for further specialised care	4	32
BASL, BSG, AUGIS²¹⁴	UK (2004)	Expert commentary and recommendations	NA	National plan for liver services in the UK	Best clinical outcomes for hepato-pancreato-biliary surgery can only be achieved at specialised centres. A planned approach is needed to develop expertise within specialised units. Many patients with liver disease are managed by GPs, but there is a gradual shift towards involving gastroenterologists and nurse specialists	4 (Guideline)	50